It is well documented that low socioeconomic couples are particularly vulnerable to experiencing relationship problems and instability. The proposed study investigates the impact of a novel relationship improvement intervention on the management of conflict in low SES couples. Building off Construal Level Theory, this research tests the hypothesis that adopting an abstract mental representation (i.e., high level construal) of one?s relationship should promote the constructive management of conflict and in turn improve long term relationship outcomes. To test the impact of a relationship construal level intervention on the management of conflict in low SES couples? everyday lives I plan to use an intensive longitudinal approach with a survey-experimental design. 146 low SES couples will be randomly assigned to either an experimental (high level construal training intervention) or control condition. The study will take place over a 3.5 month period, beginning with a 2 week long daily diary and ending with a 3 month follow up questionnaire. Couples in the experimental condition will be trained to self-induce an abstract mindset when they encounter conflict in their relationship and received regular intervention exercises throughout the study. During the diary period daily measures of conflict and relationship affect will be collected. The follow up survey will assess change in relationship quality and conflict management strategies. I hypothesize that couples in the experimental condition will display better daily conflict management (greater use of constructive conflict strategies and lower use of destructive strategies) and more conflict resolution as well as better relationship quality and conflict management at the 3 month follow-up compared to control condition couples, and improved relationship quality and conflict management compared to baseline measures. My goal is for the findings of this research to inform social service programs and policies for the promotion of relationship health and wellbeing in low SES populations.